The media’s unbalanced referendum

We now have a number of studies of how the media as a whole treated the EU referendum.

1. Levy, Aslan and Bironzo from The Reuters Institute for the Study of Journalism
   
   
2. A short piece by Deacon et al from Loughborough in this volume.
   
   
3. Cushion and Lewis from Cardiff (full paper here)
   
   
4. Moore and Ramsay from King’s College, London
   
   

The Reuters Institute study looked at the press, and after weighting for readership and visibility they found that pro-Leave articles outnumbered pro-Remain articles 68:32 (page 34). One interesting finding that I had not seen before is that voters generally split in a similar way to the balance of articles in the paper they read: the only notable exceptions were the Times (more pro-Leave articles but more pro-Remain voters) and the Mirror (more pro-Remain articles but roughly even voting split). Of course you can read this result two ways: voters were influenced by their paper or their paper reflected their reader’s views.

The King’s College study shows how the Leave campaign, through the newspapers that supported it, were able to reframe the debate on the economics of Brexit. An example that sticks in my memory was Obama’s intervention. I remember seeing an interview with a random voter asking what she thought of this, and she responded by saying how dare Obama interfere with our referendum and blackmail us over trade. It struck me as a very odd reaction at the time (particularly as Obama is popular in the UK), but of course she was simply parroting what she had read in her newspaper. The King's study clearly reveals how the Leave press used the techniques of propaganda to support their side.

The Cardiff study focused on the main news broadcasts. In contrast with the press, there was no bias in favour of Leave or Remain. However what they did find was that broadcasters essentially acted as mirrors for the two campaigns. The Remain campaign focused on Tory politicians, so the broadcasters did as well. As a result, Conservatives received much more coverage than politicians from other political parties. As the Loughborough study noted, this made the coverage ‘presidential’ in character. Journalists normally did not question statistics themselves, preferring to let the other side do any challenging. This also meant that the broadcasters focused on the details of the two campaigns, rather than providing the background information and independent assessment that many viewers clearly wanted. Rather than focus on their duty to inform, they played it safe by just letting the two campaigns do all the talking.

A consequence of the broadcast media largely providing a showcase for the politicians running the campaigns is the marginalisation of other groups, and in particular those who actually knew something about the issues being talked about. I’m not just talking about economics, but also law and international relations. The Remain campaign prefered to use international institutions (IMF, OECD etc) rather than local experts. The Leave campaign did use one academic, Patrick Minford (who figured in 90 of the articles examined by the King’s group), with the consequence that the academic economist that voters were most likely to have heard of during the campaign represented just 4% of the profession.

The danger of the broadcast media taking this approach is illustrated by the example of immigration and public services. The King’s study noted the following:

“The most consistent economic argument made by the Leave campaign – that immigration placed unsustainable pressure on public services – was frequently repeated in the editorials of some news outlets without being subject to the skeptical or forensic analysis applied to Remain’s economic arguments across the whole range of publications.”

Economists assume, for sound reasons, that in fact immigration benefits the public finances, which is one reason why the OBR thinks the deficit will be around £15 billion higher each year as a result of Brexit. So why did the Remain campaign not say this more loudly? The answer could well be because the campaign was headed by a government that had used immigration as a scapegoat for poor public services. This absence of a critique mattered a lot: at least one poll showed that the reason voters most often gave for limiting immigration was pressure on public services. Therefore by relying on the political campaigns, the broadcast media misled the public.

It is for this reason that I have argued that broadcasters should treat what an overwhelming majority of experts think are facts as facts, whatever politicians say. But I see no sign that broadcasters see the problem (with the exception of climate change), let alone have any inclination to deal with it. As far as economics is concerned, I fear the bodies that represent academic economists also want to avoid any fights. Which means that we are stuck with the status quo for some time.

I think this has a major implication for those like me who see Brexit as a huge mistake which people must be given the chance to reverse. The next few years are going to show that the many claims the Leave side made are completely false. The EU will ensure the UK is worse off as a result of leaving. Trade deals with other countries will not come to the rescue. There will be less, not more, money for public services and so on. The government’s response (unless May makes the most courageous U-turn ever) will be to wrap themselves in the flag and say that anyone who is critical is being unpatriotic, and with a large majority no effective opposition within the Conservative party will be possible. .

If Labour continues to support a Hard Brexit, all they can do is claim setbacks are the result of government incompetence. Here I disagree with Ian Dunt: criticism that takes Hard Brexit as given and just focuses on a claim that we could do the negotiation better will lose out to nationalist fervour. The reality is also that the LibDem voice is too weak, and will remain so even if they double their number of seats at the election. Given the way our media works, the only way you can constantly remind people that Brexit is a choice we could reverse is if Labour after the election adopts a much more critical position that involves support for a second referendum.

One vote that rules them all

As Justin Lewis recalls, an Ipsos Mori poll just before the EU referendum “found that while most people (70% to 17%) did not believe a claim that British people would be significantly poorer outside the EU, they were more likely to accept (by 47% to 39%) the £350m a week figure.” Such beliefs indicate both that Leave ran a much better campaign, and also that the broadcast media totally failed to inform its viewers.

Those beliefs about the economic impact of Brexit are now beginning to change, as this series of results from a different poll show:

One obvious reason for this shift is the increase in inflation that the Brexit vote has generated. This shift is important, because polls before and after the vote also suggested that a large proportion of voters only wanted to reduce immigration (as a motive for voting Leave) as long as it did not cost them any money.

But put these things together and we get something of a paradox. If being worse off was more important than reducing immigration, and more people are now convinced they will be worse off, why has popular opinion about the vote itself hardly changed. The YouGov tracker poll, which asks “In hindsight, do you think Britain was right or wrong to vote to leave the European Union?”, has hardly moved since the vote, with currently as many people saying Yes as No.

It is not just the economic data that is going the wrong way. Matthew d’Ancona quotes a senior government source as saying “the three main Brexiteers are suddenly becoming more and more vocal about the need to keep the [immigration] numbers sufficiently high for the needs of the economy.” They are right of course, but it suggests another key area in which the expectations of Leave voters will be disappointed. Not to mention the £350 million a week coming to us turning into a £50 billion bill going to the EU.

Here is a possible reason for this paradox. (I admit I have little evidence for it, and it is not the only possible explanation.) Voters feel that once a democratic decision has been made, it should be respected, even if they personally now feel less comfortable with the reasons behind the decision. It is important to respect the ‘will of the people’ for its own sake, just as it is important to keep to a contract even though you may now regret signing it. I do not think this view is sensible in this context, but that is a different issue.

You could use a similar rationalisation for Labour’s evolving attitudes to Brexit. Their latest position is that Labour will hold May to 6 tests, one of which is to “deliver the "exact same benefits" as we currently have as members of the single market and customs union”. It is of course an impossible test for May, despite what David Davis may have said. It makes no sense coming from a party that voted for triggering Article 50, unless there was some compelling reason for supporting the will of the people for its own sake.

The big question, for those like me who would much rather we stayed in the EU, is whether the same logic applies to Conservative MPs who personally favour remaining in the EU. Is there some point at which their duty to respect the vote is fulfilled, and does that point come before or after they have to ratify whatever deal May delivers? I suspect (again with not much evidence) that this depends on whether there will be a deal or not.

The logic of this suggests there will be a deal. (I wrote this before reading today’s Guardian.) Any poor poll performances from the May council elections onwards will be described as the public sending a message that there should be a deal, rather than as the public changing their mind about leaving. The tabloids will huff and puff, but May will just for once ignore them. Conservative MPs who are also Remainers will console themselves that at least disaster has been avoided. In the end, that one vote will bind us all.

Does democracy require implementing the referendum result?

We all know the EU referendum was not legally binding on parliament. That is not true of all UK referendums: the referendum on using AV did require parliament to enact whatever voters decided. Despite the lack of a legal requirement, there remains a powerful political argument that parliament was nevertheless duty bound to implement the referendum result. It is an argument that is often invoked by both government and opposition MPs. Now I have no doubt that in reality other motives are important, perhaps decisive, but because political arguments can be persuasive, it is important to debate this one.

The clearest argument along these lines comes from a post from Richard Ekins, who is a Professor of Law at Oxford University. He writes

“Parliament made clear that the decision about whether to leave the EU was to be settled by the referendum. There were good reasons, outlined above, why Parliament should not permit Brexit otherwise than by way of a referendum. Even if one denies all this, one should still accept that a referendum once held settles what should be done. For the decision to proceed thus is itself an important public decision that fairly governs how we jointly are to decide. That is, Parliament having decided to hold the referendum, and the public having participated fully in it, the result should be respected and not undone.

Political fairness and democratic principle require one to respect the outcome of the referendum even if one is persuaded that Brexit would be a very bad idea. One might think it wrong to hold the referendum, but it was held – and Parliament invited the people to decide this question. ... In short, the important constitutional question of whether Britain should remain in the EU was fairly settled by public vote.

The proposal to ignore or undo the vote is unjust. It bears noting that the relatively powerless in our polity – the poor – overwhelmingly supported exit. Ignoring the referendum would be particularly unfair to them.”

Note that this does not say that people like me should shut up about the harm that this action will cause. Instead it says that parliament, having invited people to decide, should respect that verdict. To do otherwise would be highly undemocratic, and would be particularly unjust to those who, for well known reasons, might justifiably claim that they are not well represented by the sovereignty of parliament. Arguing that the Leave campaign told lies, or that voters were deceived, does not seem to be a compelling argument against this, as exactly these charges can and are made after general elections

To help see why Ekins is wrong, it is useful to look at his discussion of the claim by Ken Rogoff that the “real lunacy of the United Kingdom’s vote to leave the European Union was not that British leaders dared to ask their populace to weigh the benefits of membership against the immigration pressures it presents. Rather, it was the absurdly low bar for exit, requiring only a simple majority.” But Ekins’ response strikes me as particularly weak. He essentially says parliament could do this because it has done it before. He goes on to say that there is “nothing at all perverse in Parliament choosing to make provision for a clear decision on point by way of a single referendum, inviting and encouraging public deliberation that culminates in a moment of clear and authoritative decision.”

This strikes me as completely ignoring Rogoff’s point. How can a 51.9% vote on one particular day represent a “clear and authoritative decision”. If a general election is close in terms of seats, that is reflected in the balance of parliament, and governments with small majorities and independently minded MPs face constraints on what they can do. What Rogoff is saying is that a referendum which only requires in theory a majority of only a single voter can never be clear and authoritative. Those who lost can justifiably claim that if the vote had been taken a day later or earlier the result could have been different, and we know they could be correct. The fact that UK governments have made this mistake in the past does not make it right. Remember we are talking about what is right politically, not what is right in law, so precedent is far less compelling.

Much the same point applies to the issue of a second referendum. He says: “Parliament having chosen already the decision-making procedure, it is not legitimate now to say that this should be set aside. The time for arguing for a two referenda requirement, or majority support in each part of the UK, was before this referendum was held.” He is certainly right that those who had won would think it is unfair to apparently change the rules of the game after the event, much as those who have lost think the whole process is deeply unfair and unjust. I also think that Ekins’ appeal to those who are otherwise unrepresented resonates with many Labour supporters, who feel that such a move would look just like the elite overriding the wishes of the people. (See Owen Jones, who questions Corbyn’s leadership but not his Brexit line.)

Except that is nonsense. If those who voted to Leave cannot get a simple majority in a second referendum when we have a lot more information about what leaving entails then that indicates something very wrong with the initial vote, and not some plot by the elite to cheat them. It is hardly undemocratic to hold a second referendum because the situation has become much clearer. As I have said before, when politicians argue that allowing a second vote is going against ‘the will of the people’ you know that you are in real trouble.

Is that unfair to one side? Of course not, because it is how politics works. Take the Scottish referendum, where Remain won by 55.4%. Just a few years later, and we could well see another referendum. To say that is different because something crucial has changed actually plays into the arguments for a second EU referendum. Unless voters perfectly anticipated the nature of the exit deal with the EU, that deal in itself is a huge and crucial change.

It seems to be neither politics nor fairness dictates that something poorly done in the past should dictate what politicians do in the future, when there is no legal constraint on them changing their minds. Holding further votes when the situation has changed cannot be undemocratic or unfair to anyone. [1]

I think all this is a useful perspective when we go back to the original question of whether parliament is obliged in some way to enact the result of the referendum we have had. Recall that Ekins says: “Parliament made clear that the decision about whether to leave the EU was to be settled by the referendum.” Now I have said in the past that I can understand why an individual MP, who has pledged to let the referendum decide their vote, should feel duty bound to keep their word. But I do question how exactly ‘parliament’ made such a pledge. An obvious way for a parliament to make such a pledge is to embed it into the terms of the referendum itself, as was done with the AV referendum. This was not done on this occasion.

It seems to me, therefore, well within the rights of any MP or Party to say that they do not regard a vote this close as binding on how they should vote. Indeed I would go further. Any MP or Party who thinks, based on the knowledge they have, that those voting Leave will over time regret their decision, has a duty to vote based on his or her judgement, rather than be tied by some vague notion around parliamentary commitment. 

But all this assumes that the Article 50 vote was just about implementing the referendum. It clearly was not just about that. Any sane discussion of the referendum has to recognise that voting Leave gave no guidance to politicians about how to leave. The referendum was not about the Single Market, the customs union etc. What the Prime Minister should have done was to allow parliament to debate the issue of how to leave, which is critical for the future of the UK. No doubt they would have given parliament a lead, but triggering Article 50 could have waited until that discussion had taken place. [2] Theresa May decided not to allow parliament that discussion.

As a result, the vote on Article 50 was not just about deciding to start the leaving process, but it also effectively became the last chance for MPs to express any view on how we should leave. That in effect made the vote a decision to leave the way May had decided, or might decide without recourse to parliament. The moment the Prime Minister did that, any obligation an MP might have felt regarding the referendum became null and void.

This is the crucial difference between 1975 and 2016, and another reason why arguments that appeal to precedent are wrong. In 1975, voters had a clear idea about what both In and Out involved. In 2016 what Leave meant was completely unclear, not least to those campaigning for it. That meant in practice that voters decided on the basis of the form of Leave they expected to happen, or perhaps were promised would happen, rather than the form of Leave the government would eventually choose.

It is for this reason that we appear to have a decidedly undemocratic result. If the referendum had set remaining against leaving for the type of hard Brexit that we are almost certain to have, it seems extremely unlikely that a majority would have voted for that. Yet those who argue that the referendum obliged MPs to vote for triggering Article 50 are in effect arguing for exactly that result. That is neither democratic, fair or indeed wise.

[1] I am sure many would argue that a referendum which came with the promise of a later referendum where you could change your mind would be too great an invitation to those who simply wanted to exercise a protest vote. I will leave that and similar arguments for others.

[2] The more people argue that such an arrangement would not have been practical, the more they illustrate how badly designed the original referendum was. Instead of debating and voting on a specific way of leaving (which could have been chosen jointly by those who wanted to leave) relative to remaining, we got a decision which was far too open ended. As a result, Leave campaigners said during the campaign that voting Leave did not imply leaving the Single Market. Once again, it seems odd to argue that parliament should not try and rectify past mistakes like this for the sake of some imagined commitment.   

Is austerity to blame for Brexit?

Mark Blyth writes

“Strip away all the electoral politics at the moment in the U.S., the U.K., Italy, Spain and elsewhere, and that's the underlying political economy. It's a creditor/debtor stand-off where the creditors have the whip hand. And yet, the more they crack the whip, the more the backlash against austerity, in all its forms, gains strength.”

Or in other words, it is all about austerity. That is a big claim, particularly when applied to the current US elections, but I want to examine it in the specific case of the EU referendum. In short, did austerity cause Brexit? Given how opposition to austerity has been such an important part of this blog, in some ways it is an attractive line for me to take, but I do try and base what I say on evidence rather than on what is convenient.

In the past I’ve argued that there is a massive problem with this idea, and the related idea that Brexit was a more general protest vote against elites. The obvious time to protest against austerity was the 2015 General Election. Yet rather than protest against the party that introduced austerity and promised much more of it, the British people gave the Conservatives a surprise victory.

It is nevertheless possible to argue that austerity caused Brexit in more subtle ways. I’ve also argued in the past that some of the concern over immigration is actually the result over concern about reduced public services and low wages, and a belief that the issues are linked. To the extent that reduced access to public services and to some extent low wages is actually the result of austerity, and if much of the public believe that austerity is nevertheless necessary, then what should be a protest over austerity could get displaced as a protest about immigration.

If your response to this idea is to say that concern over immigration is also a result of racism and xenophobia, I would agree, but argue that this is beside the point. When talking about the Brexit vote, we should be concerned about what you might call the swing voters, a point that Chris Dillow also makes. Remember that a large number of those voting Leave would not have been prepared to pay anything to reduce immigration: they do not sound like voters whose overriding concern is to see less foreigners on their streets. It is the Brexit voters who thought Brexit would make them better off that we should be concerned about.

So there is a possible mechanism by which austerity could have caused Brexit. That mechanism is part of a more general phenomenon: when things get tough, people become much more receptive to potential unfairness. It is what helps drive a belief that welfare goes to scroungers, a belief that some Conservative ministers seem happy to encourage.

Is there any evidence to support the idea that the mechanism I’ve outlined was important? Here is polling on the EU referendum over a long time period (source).

If there is a general recession effect here (i.e. applying to every UK recession) it is masked by other factors. However after the global financial crisis and austerity we did see a big shift against the EU, although that could also be explained by the Eurozone crisis. The drop in support for leaving before the 2015 election seems to go against the austerity caused Brexit hypothesis, but that was also a time the government and much of the media was claiming that the UK economy was recovering strongly. The true state of the NHS only became apparent to most people after the election, when support for Leaving revived. So I do not think this historical evidence is conclusive either way.

There is some econometric evidence for a link between the extent of public service cutbacks and the proportion of people voting Leave (for a summary, see this VoxEU article). But as the article itself notes, the measure of austerity used could simply be acting as a proxy for more long term deprivation, which is a widely acknowledged influence on Brexit.

In my view more compelling evidence for an impact of austerity on the Leave vote comes from the little polling evidence we have on why people think high levels of EU immigration is a problem. Here is the result of a poll on this that I have reproduced before, which speaks for itself.

As I have argued here, the current squeeze on the NHS is unprecedented. The share of NHS spending in GDP has a natural tendency to rise over time, for reasons that are well understood. Yet not only did few in the media contest the ‘common sense’ idea that austerity was necessary, but also voters hear time and again that NHS spending is being protected. As they see services deteriorate, it is not surprising that they conclude that there is just too much demand. I doubt very much that it is coincidence that the Leave campaign’s bus had the £350 million a week going to the NHS.

So there is evidence that links austerity to the Brexit vote, particularly if we remember we are not talking about a core vote that would have voted Leave anyway (because of xenophobia, for example), but the swing voters who at other times and circumstances might have voted the other way. But what about arguments that the Leave vote reflected a reaction to deprivation caused by globalisation, or that it was the result of the malign influence of the tabloid press? (I’ve made both arguments in the past.)

There is some complementarity here. As I noted at the beginning, the media is important in transforming concern about austerity from the politicians that impose it into concern about immigration. More importantly, there is no need to find the cause of Brexit. It seems quite possible to believe that the vote would have gone the other way if we had not had austerity, or if we had a tabloid press that was not just a cheerleader for Leave and a broadcast media indifferent to expertise, or if the impact of globalisation had been offset in various ways. With the vote so close, it is legitimate to argue that all three on their own might have been responsible for Brexit, or equivalently that the Brexit result was a consequence of a perfect storm of bad policies and institutions.